Use of Fin Rays and Fin Spines in Nonlethal Age Estimation of Florida Bass

Unbiased and precise estimation of age is critical to understanding processes related to the dynamics of fish populations, including recruitment, age structure, mortality, and growth (Lai and Gunderson 1987; Kimura and Lyons 1991; Campana 2001). Reliable estimation of age depends on the correct identification and interpretation of annuli formed in anatomical structures (Beamish and McFarlane 1983; Campana 2001; Quist et al. 2012). Many hard structures have been used to estimate ages of fishes, including scales, otoliths, fin rays, fin spines, cleithra, and opercles, but precision and bias in estimated ages vary with species and structure used (Chilton and Beamish 1982; Dutka-Gianelli and Murie 2001; Whitledge 2017; GSMFC and ASMFC 2020).

Florida Bass Micropterus salmoides (Kim et al. 2022) are assigned ages via the number of annuli counted in their sagittae, a technique that has been validated in the species to age 5 (Hoyer et al. 1985) and in Largemouth Bass Micropterus nigricans, to age 16 (Buckmeier and Howells 2003), but removal of otoliths requires killing the fish. Bass that attain trophy size (e.g., ≥ 3.63 kg) are infrequent in targeted sampling for their species, so sacrificing the few large bass that are captured can be undesirable; therefore, these rare individuals are often excluded from or underrepresented in age samples (Crawford et al. 1996; Dutterer et al. 2014; Klein et al. 2017). This can lead to biased estimates of population parameters or missed opportunities for obtaining information about the largest individuals. Effective nonlethal aging methods could alleviate problems caused by missing data while providing benefits for fishery scientists tasked with managing trophy bass, fish populations in water bodies undergoing restoration, or data collected from tournaments and citizen scientists. These methods would also add a new dimension to ongoing field studies and experiments by allowing biologists to include estimated age as a factor that influences growth, mortality, and behavior without sacrificing fish.

Nonlethal aging methods also align with the widespread adoption of catch and release practiced by most bass anglers, who place high priority on the live release of all sizes of bass (Allen et al. 2008). Historically, bass fisheries were harvest oriented, but catch-and-release angling increased in popularity in the 1980s and transformed the character of the fisheries (Allen et al. 2008; Dotson et al. 2013; Long et al. 2015; Taylor et al. 2019). Most Florida Bass fisheries are characterized by high voluntary release rates (77%–95%; A. C. Dutterer, Florida Fish and Wildlife Conservation Commission [FWC], unpublished data) and can produce bass weighing ≥ 3.63 kg. Stakeholders place high value on trophy bass, which is reflected by numerous fishery scientists who have created programs like TrophyCatch and ShareLunker to collect catch information about trophy bass that informs research projects and ultimately leads to management actions (Myers and Allen 2005; FWC 2011; Dutterer et al. 2014). Thus, a reliable, nonlethal aging method for Florida Bass would allow fishery scientists to adopt sampling protocols that are better aligned with stakeholders’ wants and values while decreasing mortality in trophy bass.

Fin rays and fin spines can be collected without sacrificing the fish, and their annuli have been used to age individuals of freshwater and marine fish species (Beamish 1981; Herbst and Marsden 2011; Baker et al. 2018). Age estimates using Largemouth Bass pectoral rays, dorsal spines, and anal spines have been assessed at several locations with varied results. When known-age Largemouth Bass were used in Klein et al.’s (2017) study, age estimates from dorsal- and anal-spine sections were less accurate (percent agreement [PA, Sikstrom 1983] = 34% and 12%, respectively) than were those from sagittae sections (PA = 92%). Poor precision between readers and bias in Largemouth Bass fin ray and spine age estimates have also been documented relative to age estimates from sagittae (Maraldo and MacCrimmon 1979; Sotola et al. 2014; Blackwell et al. 2019; Griffin et al. 2022). A few studies, however, have demonstrated high between-reader precision of age estimates for Largemouth Bass using fin structures (Morehouse et al. 2013; Klein et al. 2017; Griffin et al. 2022). Yet, the results of the earlier studies on Largemouth Bass could differ for the faster-growing Florida Bass.

To our knowledge, the use of anal, dorsal, and pelvic rays and pelvic spines have never been assessed for age determination in Florida Bass or other black basses, nor have growth zones in their fin structures been validated as annuli (Maceina et al. 2007; Spurgeon et al. 2015). Understanding the bias and precision of age estimates from nonlethally removed fin structures and the variables that affect these metrics will offer new results and management value for Florida Bass. We evaluated the use of anal-, pelvic-, dorsal-, and pectoral-fin rays and anal-, pelvic-, and dorsal-fin spines in estimating the age of Florida Bass. Our objectives were 1) to evaluate the difference in age estimates from fin structures compared to age estimates from sagittae (i.e., bias), 2) to assess the absolute difference in age estimates from each aging structure made by two readers between two aging sessions (i.e., precision), and 3) to indirectly validate annulus formation in a select Florida Bass fin structure by marginal increment analysis and compare timing of annulus formation to that of sagittae.

We collected Florida Bass of 90–600 mm maximum total length monthly from February 2017 through January 2018 from Rodman Reservoir in Putnam and Marion counties, Florida, using daytime boat electrofishing (n = 686). We collected all Florida Bass incorporated in the precision and bias analyses during March and April (Table S1, Supplemental Material). We weighed (g) and assigned a unique identification code to the fish (Fish ID) and then brought them to the laboratory for dissection. We extracted, cleaned, and air-dried the sagittae from each fish and mounted the right otolith (n = 275) onto a microscope slide with superglue for sectioning. We then used a low-speed IsoMet™ saw (Buehler, Lake Bluff, IL) fitted with three diamond-wafering blades (Buehler, Lake Bluff, IL) separated by two 0.5-mm stainless steel spacers to yield two transverse 0.5-mm sections from the core of each otolith. We permanently mounted the sections onto microscope slides using Histomount^® (National Diagnostics, Atlanta, GA; Murie et al. 2009).

We removed the pectoral, anal, dorsal, and pelvic fins from Florida Bass using multipurpose snips. We tested two techniques of cutting fins to verify that nonlethal fin clips (clipped no deeper than flush with the fish’s body) contained the same number of annuli as fins removed in whole through cutting into the body. We removed fins from a subsample of Florida Bass (n = 36) by cutting into the body, whereas we cut the remainder of fins no deeper than the point at which they joined the fish’s body (n = 650). We stored fins individually in a freezer and thawed them before cleaning. We dipped the fins in 95°C water for 1–2 minutes to ease removal of adipose, connective, and dermal tissue (Murie et al. 2009). We excised rays (made up of two hemitrichs) 3–5 from the detached pectoral, dorsal, and anal fins; rays 2–4 from pelvic fins; and spines III–V from dorsal fins (n = 85). We also excised the pelvic-fin spine and anal-fin spine III (n = 47 and 49, respectively). We stored the excised fin structures in labeled coin envelopes with their basal ends exposed to the air (Chilton and Beamish 1982). Following desiccation in a drying box with a fan, we embedded the fin structures in polyester thermoplastic resin (LOCTITE®, Henkel, Düsseldorf, Germany; Murie et al. 2009) and took three to four cross sections (1.0 mm thick) along the length of the embedded fin structure using a low-speed saw fitted with a diamond wafering blade. We permanently mounted the sections on semi-frosted microscope slides using Flo-Texx® (Lerner Laboratories, Pittsburgh, PA; Murie et al. 2009).

We examined otolith sections with a dissecting microscope (25–40×) and transmitted light. We viewed fin-structure sections with a compound microscope (25–100×) fitted with a green narrowband-interference filter (540 nm; Murie et al. 2009; GSMFC and ASMFC 2020) using transmitted light. We considered an annulus to be a combination of an opaque plus a translucent band. We reviewed each section for annulus uniformity (number) and clarity as we moved distally from the base along the fin structure. We also compared the number of annuli present in each structure between the two cutting techniques to determine if the first year of growth was included. We used a microscope camera to photograph structure sections (Figure 1) using either QCapture Pro 7 (Teledyne Qimaging, Surrey, British Columbia, Canada) or AmScope™ 3.7 (United Scope LLC, Irvine, CA) software to display the digital photographs. We assumed a January 1 birth date and that the age estimate from the otolith was the true age for all Florida Bass (Taubert and Tranquilli 1982; Hoyer et al. 1985; Crawford et al. 1989; Buckmeier and Howells 2003). Prior to estimating ages in the study, we trained on a small reference set of fin-structure sections based on the otolith-assigned age (n = 41; 1–7 years).

Readers one and two twice read the clearest section of the otolith, fin ray, and spine and assigned an age to the Florida Bass from which the structure was taken. We did not include these fish in the reference set. If our individual age estimates across our two aging sessions did not agree, we aged the sample a third time to designate a concert age. We estimated the age of each fish using its otolith by counting the opaque bands (i.e., slow growth) of fully formed annuli. We assigned a code to the translucent band on the otolith margin based on its width relative to that of the previous complete (fully formed) annulus (Quist et al. 2012; GSMFC and ASMFC 2020). We assigned margin codes as: 0 = only the opaque band was visible at the margin; 1 = the translucent band was one-quarter as wide as that in the previous complete annulus; 2 = the translucent band was half as wide as that in the previous complete annulus; 3 = the translucent band was three-quarters as wide as that in the previous complete annulus; and 4 = the translucent band was the same width as that in the previous complete annulus. In our age estimates, margin codes of 3 or 4 were counted as a year. We estimated the age of each fish using its fin structures by counting the translucent bands (i.e., slow growth) and assigned an established code to the opaque band on the margin relative to that of the previous complete presumed annulus (presumed because yearly growth has not been validated; Chilton and Beamish 1982; Murie et al. 2009; GSMFC and ASMFC 2020).

We used two different metrics to understand discrepancies in age estimation of Florida Bass. We measured the precision of age estimates by calculating the absolute difference between replicate reads of each aging structure for each reader (Table S2, Supplemental Material). We measured bias by calculating the difference between fin structure age estimates and the corresponding concert otolith age estimate for the Florida Bass from which the fin structure was taken. Thus, our measure of bias is relative to otolith age. We used raw difference rather than absolute difference because unlike precision, the direction of the difference is meaningful for assessing bias. When calculating bias, we used each of our concert ages for each structure rather than our ages assigned from both aging sessions, thus we had one observation per structure per reader, rather than two.

We used an information theoretic approach to evaluate the effect of structure, reader, and age on precision and bias of age estimates. We developed two candidate model sets, one for precision and one for bias. We used linear mixed-effects models because it allowed us to account for non-independence among multiple estimates derived from structures taken from the same individual fish. Fish ID was included as a random effect in each model in both model sets. We also included three single fixed-effect models (structure, reader, age) as well as multiple models that contained different combinations of these predictors and their interactions as fixed effects in each model set. We fit precision models using the absolute difference between replicate age estimates as the dependent variable. We fit the bias models using the difference between fin structure age estimates and otolith age estimates as the dependent variable.

We ranked models using Akaike’s Information Criterion corrected for small sample size (Akaike 1973; AICc; Hurvich and Tsai 1989). We considered any model with a ΔAICc of less than five plausible. We conducted all analyses with R (R Core Team 2022). We fit linear-mixed effects models using the lme4 package (Bates et al. 2015). We generated Akaike Information Criteria tables using the bbmle package (Bolker and R Development Core Team 2022) and additional descriptive statistics (coefficient of variation, CV, Chang 1982; average percent error, APE, Campana 2001) using the FSA package (Ogle et al. 2018) and custom R functions.

We also conducted a marginal increment analysis for otoliths from the age-4 cohort of Florida Bass (n = 80) and compared plots of C by month between dorsal spines and otoliths to determine the difference in time of year of growth-zone deposition between the two structures (Dutka-Gianelli and Murie 2001).

The top-ranked model to evaluate precision included the fixed effects of age, structure, and the age*structure interaction, and the random effect of Fish ID (Table 1). There were no competing models. The difference between replicate age estimates increased with age but varied by structure (Figure 2). The model predicted lesser differences between replicate age estimates from otoliths than from all the assessed aging structures (Figure 2). Replicate age estimates from pelvic spines produced the greatest predicted differences, which were more influenced by age than other structures (Figure 2). All other fin structures provided similarly precise age estimates to one another based on widely overlapping confidence intervals (Figure 2).

The most strongly supported model to evaluate bias in estimated ages from fin structures included the fixed effects of age, reader, and the age*reader interaction, and the random effect of Fish ID (Table 2). There were no competing models. The model that included only structure as a fixed effect was less supported than the null model (Table 2). As the age of the Florida Bass increased, fin structure age estimates were more likely to underestimate the otolith age (Figure 3); however, the strength of the age effect varied by reader (Figure 4). Reader two displayed a greater difference in age estimates between fin structures and otoliths for older fish compared to reader one (Figure 4).

Marginal increment analysis indicated that annual formation of growth zones occurs in dorsal spines from Florida Bass aged 3, 4, 5, and 6 years. Annulus formation in Florida Bass dorsal spines began in the fall from November to March (C minimum 21% in November; Figure 5). Whereas C derived from otoliths was at its minimum, 50%, in April; thus, annulus formation was completed in the spring from April to June (Figure 5; Table S3, Supplemental Material). Annulus formation was completed in Florida Bass dorsal spines before that of otoliths.

In this comprehensive evaluation of eight Florida Bass ageing structures, we identified that precision of age estimates was influenced by the age of the Florida Bass. Otoliths were least affected and pelvic spines were the most affected aging structure. We found little difference among fin structures in bias when compared with the otolith age, but age estimates for older Florida Bass were more biased than for younger Florida Bass. Bias varied among readers, which suggests that reader experience or skill may influence the utility of fin structure age estimates. Regardless of the fin structure evaluated, most age estimates were within a year of the otolith age, suggesting that most fin structures may have utility for estimating the age of Florida Bass in a nonlethal manner (Table S4, Supplemental Material). For future age estimation of fin structures in Florida Bass, we recommend that readers 1) repeatedly examine a subsample of fin structures taken from Florida Bass of various known ages prior to estimating ages, 2) first practice estimating age in younger Florida Bass, during which they will learn to recognize the first annulus, and 3) compare images of fin structures among members of the same cohort to help decrease bias.

Dorsal spines became our preferred fin structure around which to develop nonlethal age estimation techniques. It was easier to excise, clean, and section dorsal spines than the other fin structures. It was unnecessary to cut into the body of the fish when excising fin structures, as the structures were ageable (e.g., all annuli retained) when the fins were cut as close to the body as possible. Experimental evidence verified that dorsal spines removed close to the body did not affect survival of Florida Bass (Lindelien et al. 2021). The number of annuli were uniform across the three to four sections taken distally along the spine, but we observed higher contrast in annuli sectioned closer to the base. Basal sections taken from Lutjanidae dorsal spines provided the most comparable age estimates to those of otoliths, whereas sections from the middle and distal areas along the spine resulted in underaged old fish (Baker et al. 2018). We recommend using sections taken from the proximal area of the dorsal spine to estimate age.

Marginal increment analysis indicated that annuli are formed in Florida Bass dorsal spines although formation occurred earlier than for otoliths (e.g., November to March versus April to June). Timing of annulus formation has previously been evaluated for Florida Bass, and our results agree with this study where annuli were formed from April to July (Crawford et al. 1989). It would have been difficult to assign individuals into correct age classes without conducting the marginal increment analysis on dorsal spine sections, as we had no previous knowledge that differences in timing of annuli formation occurred (GSMFC and ASMFC 2020). Campana (2001) advised using one cohort in marginal increment analysis. However, our sample did not have enough individuals from any one cohort to support a single-cohort approach, so we included Florida Bass aged 3-6 to bolster monthly sample size. Therefore, our results should be interpreted cautiously and could be improved with increased sample sizes of Florida Bass from the same cohort. We plan to directly validate annuli in dorsal spines from Florida Bass using hatchery-raised fish of known ages (Campana 2001; Spurgeon et al. 2015).

We were better able to identify and visually discriminate between annuli in dorsal-spine sections than other fin structures because they had the largest cross-sectional area and less compacted translucent bands at the margin. For age-1 and age-2 Florida Bass, we suspect that we occasionally misidentified the core as the first annulus, as the outside of the core resembled a band, and false bands were also common inside the core. For a few age-6+ Florida Bass, the first annulus in dorsal-fin-spine sections was occluded because of the expansion of the vasculature-filled lumen throughout the core causing erosion of the first translucent and opaque bands and likely resulted in under-aging (Figure S1, Supplemental Material). Many studies have documented similar occurrences associated with early growth in the use of fin structures for age estimation (Beamish 1973; Donabauer 2010; Isermann et al. 2010; Elzey and Trull 2016; Dembkowski et al. 2019; Carroll et al. 2023).

Dorsal spines have been assessed for estimating the age of Largemouth Bass in Canada (Maraldo and MacCrimmon 1979), Minnesota (McInerny et al. 2017), New York (Sotola et al. 2014), Indiana (Carnahan 2009), South Dakota (Blackwell et al. 2019), Georgia (Klein et al. 2017), and Oklahoma (Griffin et al. 2022), and, generally, estimates were inconsistent and not comparable to those made using otoliths. Recently, Griffin et al. (2022) estimated ages using whole dorsal spines, achieving low agreement with age estimates from otoliths (APE = 32.25%). On the contrary, Zhu (2015), used sections from the third dorsal spine to successfully estimate the ages of Largemouth Bass ≤ 7 years and Smallmouth Bass Micropterus dolomieu ≤ 9 years. Though bias in dorsal spine age estimates was no different than other fin structures based on our models, we produced a lower mean CV and APE across readers (8.44% and 5.97%, respectively; Table S4, Supplemental Material) between dorsal spine and otolith age estimates compared to the aforementioned studies. Using a modeling approach allowed us to capture the multiple sources of variation (i.e., age and reader) more accurately in the dataset than with descriptive statistics alone. To our knowledge, most other studies regarding Largemouth Bass age estimates have used descriptive statistics to evaluate the use of aging structures, which we do not find useful for assessing precision and bias. McBride (2015) demonstrated that PA can vary considerably for simulated age estimates that are precise and unbiased, suggesting that comparing between studies and making inferences from this index alone is not ideal. We reported PA, CV, and APE to provide reference to other study results, wherein CV and APE are most accepted in the literature (Campana 2001).

Some of the other disparity in previous studies of estimating age using Largemouth Bass dorsal spines likely results from processing methods. For instance, Klein et al. (2017) collected the first dorsal spine, which is the smallest and more apt to be damaged than more posterior spines (GSMFC and ASMFC 2020). Maraldo and MacCrimmon (1979) and Sotola et al. (2014) also collected the first dorsal spine, but neither reported the width of the sectioned spines they used for age estimation. At first, in the present study, we examined sections of the first and second dorsal spines to find that their bands were crowded and overlapped, especially at the margin; therefore, we did not use these spines in our study. We recommend collecting the central-most spines, which are the longest or largest, and then taking 0.5–1.0-mm-thick sections to reveal the greatest surface area and annuli that are easily distinguished and not crowded (Whitledge 2017; Lindelien 2018).

Our study provided insight into visual differences in annuli among fin structures in general and across fin structures from individual Florida Bass. Anal- and dorsal-ray sections, for example, often contained multiple connected, translucent bands that readers often counted as two instead of one (Kocovsky and Carline 2000); this led to age overestimation unless distinct separation between the bands was recognized (Figure S2, Supplemental Material). Pelvic-ray sections displayed distinctive presumed annuli, and even bands that formed in multiples were usually identifiable on the larger hemitrich, in which the bands were less compacted than in the smaller hemitrich (Figure S3, Supplemental Material). These differences in patterns of presumed annuli across structures should be addressed during training and be incorporated into age-estimation protocols.

To our knowledge there are no published studies on Micropterus species that assessed the aging utility of pelvic rays and spines, dorsal rays, or anal rays, but a few have evaluated the age estimates from pectoral rays compared to otoliths. For Smallmouth Bass younger than age-5, Rude et al. (2013) recommended using pectoral rays for age estimation by experienced readers. However, Maraldo and MacCrimmon (1979) did not consider pectoral rays useful for determining the age of Largemouth Bass. We were able to identify presumed annuli in both hemitrichs, and once we recognized the core, the first band was wide and evident in pectoral-ray sections (Johnson and Weston 1995). The mean CV across readers between replicate pectoral ray age estimates was 7.61% (Table S5, Supplemental Material) which is lower than earlier studies in which average coefficient of variations (ACVs) were 12.85% (Rude et al. 2013), 21.70% (Morehouse et al. 2013), and 18.68% (Griffin et al. 2022). This suggests that pectoral rays can yield reasonably precise age estimates, however, as with the other fin structures we evaluated, we found that precision was inversely related to fish age.

Several studies have also deemed anal spine age estimates for Largemouth Bass as inaccurate (Klein et al. 2017; ACV = 19.00%) and inferior to those estimated from otoliths (Blackwell et al. 2019). Griffin et al. (2022) reported an average percent error (APE, Campana 2001) of 40.08 between otolith and anal spine age estimates, whereas our APE was 5.24%. We do not recommend using anal and pelvic spines to age Florida Bass. These fin structures were small and short, making them difficult to clean and section. Once processed, translucent and opaque bands were inconsistent in width and often indiscernible in anal and pelvic-spine sections. For instance, in pelvic-spine sections, the first translucent bands were wide and formed as multiple joined bands among false bands (non-yearly growth), whereas presumed annuli at the margins were compacted (Figure S4, Supplemental Material). We collected Florida Bass during March and April for our analyses on precision and bias to align with spring sampling and aging that occurs for Rodman Reservoir. According to Crawford et al. (1989) it is more difficult to interpret the otolith margin from April through June in Florida Bass. Estimating ages from Florida Bass fin structures in months other than March and April could help assess if bias and precision are constant throughout the year.

The use of nonlethal age estimation will enhance Florida Bass research, management, and conservation by providing estimates of population parameters without sacrificing fish. This method would increase the likelihood that trophy Florida Bass (> 3.63 kg), which some managers are reluctant to sacrifice, would be included in age and growth assessments. This could make estimates used to assess populations more robust by including more ages from the largest, and likely oldest, individuals of a population (Dutterer et al. 2014). Although our results indicate greater bias in age estimation among the oldest Florida Bass in our study, our technique offers opportunities for use with trophy bass because of their value in Florida Bass fisheries. In considering using fin structures for age estimation, managers must carefully consider the tradeoffs among data gaps, data accuracy, and the acceptability of removing valuable individuals from populations. We aim to help managers better understand these tradeoffs in a future publication by simulating the effects of our empirical age estimation errors on several metrics commonly used in fisheries stock assessments.

Nonlethal age estimation techniques could also be valuable in various management scenarios in which sacrificing bass is undesired, such as for small populations, populations that are closed to harvest, and populations that are intensively managed for trophy fish. Furthermore, using fin structures for age estimation could enhance management and conservation of rare, endemic Micropterus spp. (Quist et al. 2012; King et al. 2018) such as Choctaw Bass M. punctulatus, Bartram’s Bass M. sp. cf. cataractae (Taylor et al. 2019), and Shoal Bass M. cataractae (Lee et al. 1980). The use of nonlethal structure-removal methods could also be integrated into fishery-dependent data sources that incorporate live release. Such situations include tournaments or citizen-science programs, like the FWC’s TrophyCatch (Dutterer et al. 2014) where more trophy Florida Bass are encountered than in fishery-independent sampling (Dutterer et al. 2014; Hall et al. 2019).

Table S1. We collected Florida Bass Micropterus salmoides using daytime boat electrofishing during February 2017–January 2018 from Rodman Reservoir in Putnam and Marion counties, Florida to assess the utility of seven fin structures for nonlethal aging. We assigned each fish a unique identification code (Fish ID), maximum total length (mm), sex (f = female, m = male), reference set code (y = used for training, n = not used for training), and we removed sagittae, dorsal-fin rays, dorsal-fin spines, pectoral-fin rays, pelvic-fin rays, pelvic-fin spines, anal-fin rays, anal-fin spines (y = collected from fish, n = not collected from fish) from each fish. UFFLMB-47 was thrown out during processing.

Table S2. We removed Florida Bass Micropterus salmoides sagittae as well as pectoral, anal, dorsal, and pelvic fins flush with the body during 2017 to 2018. We excised and embedded the middle rays or spines from each fin and took three to four cross sections (1.0 mm thick) along the length of the fin structure using a low-speed saw and diamond wafering blade. We took two 0.5 mm thick cross sections from the otoliths using two 0.5 mm metal spacers and three diamond wafering blades. We (reader one = SL, reader two = DP) permanently mounted the sections and assigned ages to each Florida Bass’ Micropterus salmoides eight aging structures during two replicate aging sessions to assess precision of age estimates in a nonlethal ageing study. We recorded unique identification code (Fish ID), reader, aging structure type, first age (years), second age (years), difference between first and second age (years), and concert age estimates for each reader (years).

Table S3. We obtained measurements for 1 year of growth using marginal increment analysis of sagittae and dorsal spine sections taken from Florida Bass Micropterus salmoides collected from Rodman Reservoir, Florida during February 2017 to January 2018. We used this technique to indirectly validate annulus formation for dorsal spines from fish aged 3, 4, and 5 years by identifying the minimum mean growth increment (i.e., smallest index of completion) equating to the timing of annulus formation. We compared the timing of annulus formation in dorsal spines to that of otoliths from the age-4 cohort. We recorded the unique identification code (Fish ID), type of aging structure, month the bass was collected, cohort for fish (years), width of ultimate zone of growth in structure (mm, Wn), width of the penultimate zone of growth (mm, Wn – 1), and index of completion (mm, C = [Wn∕Wn – 1] × 100).

Table S4. We removed Florida Bass Micropterus salmoides sagittae as well as pectoral, anal, dorsal, and pelvic fins during 2017 to 2018. We excised and embedded the middle rays or spines from each fin and then took three to four cross sections (1.0 mm thick) along the length of the fin structure using a low-speed saw and diamond wafering blade. We took two 0.5 mm thick cross sections from the otoliths using two 0.5 mm metal spacers and three diamond wafering blades. We (reader one = SL, reader two = DP) permanently mounted the sections and assigned ages to each Florida Bass’ Micropterus salmoides seven fin structures and to each bass’ sagittae to assess bias of age estimates in a nonlethal aging study. We recorded a unique identification code (Fish ID), reader, fin structure type, fin structure age (years), otolith age (years), and difference between fin structure and otolith age (years).

Table S5. We collected Florida Bass Micropterus salmoides from Rodman Reservoir, Florida for a nonlethal ageing study from February 2017 to January 2018. We collected and cross-sectioned sagittae and fin structures to estimate the ages of the fish. We used a compound microscope (25–100×) fitted with a green narrowband-interference filter and transmitted light to estimate fin structure ages. We used a dissecting microscope (25–40×) and transmitted light to estimate otolith ages. We calculated metrics of precision for age estimates from two replicate aging sessions averaged across two readers for all aging structures. We recorded the eight aging structures, structures aged (n), R = reader, PA = percent agreement (%), ± 1 yr = PA within a year of the exact age (%), ± 2 yr = PA within two years (%), ± 3 yr = PA within three years (%), ACV = average coefficient of variation, and APE = average percent error.

Figure S1. We removed and cross-sectioned Florida Bass Micropterus salmoides dorsal-fin-spines (1.0 mm thick) collected from Rodman Reservoir, Florida during February 2017 to January 2018. Dorsal-fin-spines were one of seven fin structures evaluated for nonlethal age estimation. Dorsal-fin-spine sections displayed a vasculature-filled lumen that at times expanded throughout the core and eroded the first annulus. The section pictured in the left frame of the photo is nonannotated whereas the section pictured in the right frame of the photo displays white points which indicate annotated translucent bands.

Figure S2. We removed and cross-sectioned Florida Bass Micropterus salmoides anal-fin-rays (1.0 mm thick) collected from Rodman Reservoir, Florida during February 2017 to January 2018. Anal-fin-rays were one of seven fin structures evaluated for nonlethal age estimation. Anal-fin-rays contained multiple connected translucent bands estimated as being formed in one year of growth.

Figure S3. We removed and cross-sectioned Florida Bass Micropterus salmoides pelvic-fin-rays (1.0 mm thick) collected from Rodman Reservoir, Florida during February 2017 to January 2018. Pelvic-fin-rays were one of seven fin structures evaluated for nonlethal age estimation. Pelvic-fin-ray sections included, right, the larger hemitrich and, left, the smaller hemitrich.

Figure S4. We removed and cross-sectioned Florida Bass Micropterus salmoides pelvic-fin-spines (1.0 mm thick) collected from Rodman Reservoir, Florida during February 2017 to January 2018. Pelvic-fin-spines were one of seven fin structures evaluated for nonlethal age estimation. Pelvic-fin-spine-sections displayed inconsistent widths for translucent and opaque bands.

Reference S1. Carnahan DP. 2009. An aging comparison between otoliths, scales, and dorsal spines from Patoka Lake Largemouth Bass. Indianapolis: Indiana Department of Natural Resources. Crawford, Dubois, Orange Counties Fish Management Report.

Reference S2. Donabauer SB. 2010. Comparing otoliths, dorsal spines, and scales to estimate age, growth, and mortality between male and female Walleye from Brooksville Reservoir, Indiana. Indianapolis: Indiana Department of Natural Resources. Final Report/.

Reference S3. [FWC] Florida Fish and Wildlife Conservation Commission. 2011. Black bass management plan (2010–2030). Tallahassee, Florida: FWC.

Reference S4. [GSMFC and ASMFC] Gulf States Marine Fisheries Commission and Atlantic States Marine Fisheries Commission. 2020. A practical handbook for determining the ages of Gulf of Mexico and Atlantic Coast fishes, 3^rd edition. Ocean Springs, Mississippi and Arlington, Virginia: GSMFC and ASMFC. Publication 300.

Reference S5. McInerny MC, Stewig JD, Hodgins NC, Hoxmeier RJH. 2017. Evaluation of age estimates made with scales, dorsal spines, and whole otoliths as surrogates for ages estimated with halved otoliths of Smallmouth Bass and Largemouth Bass in Minnesota. Minnesota Department of Natural Resources. Investigational Report 567.

Reference S6. Chilton DE, Beamish RJ. 1982. Age determination methods for fishes studied by the groundfish program at the Pacific Biological Station. Nanaimo, B.C.: Department of Fisheries and Oceans Resource Services Branch. Canadian Special Publication of Fisheries and Aquatic Sciences 60.

We thank the U.S. Fish and Wildlife Service’s Sport Fish Restoration Program for providing Federal Aid funding and the University of Florida for providing personal support and tuition (S. Lindelien) during the first year of the study. We thank J. Dotson and E. Nagid for their guidance. T. Van Natta assisted in fish collections. M. Hoyer and E. Camp provided insight about early drafts of the manuscript. D. Murie and her lab provided insight into processing and aging structures and marginal increment analysis at the beginning of the project, as well as use of a sectioning saw, laboratory space, and materials for processing fin structures. B. Crowder, K. Bonvechio, and Journal of Fish and Wildlife reviewers and editors provided editorial guidance that improved the organization and clarity of this manuscript.

Any use of trade, product, website, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government.